Allen: A high level trigger on GPUs for LHCb

Abstract

We describe a fully GPU-based implementation of the first level trigger for the upgrade of the LHCb detector, due to start data taking in 2021. We demonstrate that our implementation, named Allen, can process the 40 Tbit/s data rate of the upgraded LHCb detector and perform a wide variety of pattern recognition tasks. These include finding the trajectories of charged particles, finding proton-proton collision points, identifying particles as hadrons or muons, and finding the displaced decay vertices of long-lived particles. We further demonstrate that Allen can be implemented in around 500 scientific or consumer GPU cards, that it is not I/O bound, and can be operated at the full LHC collision rate of 30 MHz. Allen is the first complete high-throughput GPU trigger proposed for a HEP experiment.

12 pages, 12 figures, 2 tables

References in corpus (2)

• A comprehensive real-time analysis model at the LHCb experiment
• A parallel-computing algorithm for high-energy physics particle tracking and decoding using GPU architectures

Cited by in corpus (34)

• Feebly-Interacting Particles:FIPs 2020 Workshop Report
• More Indications for Lepton Nonuniversality in $b\to s{\ell }^{+}{\ell }^{-}$
• The LHCb upgrade I
• Unleashing the full power of LHCb to probe Stealth New Physics
• GPU-accelerated machine learning inference as a service for computing in neutrino experiments
• A Comparison of CPU and GPU implementations for the LHCb Experiment Run 3 Trigger
• GPU coprocessors as a service for deep learning inference in high energy physics
• High-density gas target at the LHCb experiment
• LHCb potential to discover long-lived new physics particles with lifetimes above 100 ps
• Studying the potential of Graphcore IPUs for applications in Particle Physics
• A fast and efficient SIMD track reconstruction algorithm for the LHCb Upgrade 1 VELO-PIX detector
• A GPU-based Kalman Filter for Track Fitting
• Robust and Provably Monotonic Networks
• Real-time data processing with GPUs in high energy physics
• A quantum algorithm for track reconstruction in the LHCb vertex detector
• Offloading electromagnetic shower transport to GPUs
• Graph Neural Network-Based Track Finding in the LHCb Vertex Detector
• Portable acceleration of CMS computing workflows with coprocessors as a service
• Data Flow in the Mu3e DAQ
• FunTuple: A new N-tuple component for offline data processing at the LHCb experiment
• Summary of the trigger systems of the Large Hadron Collider experiments ALICE, ATLAS, CMS and LHCb
• Looking Forward: A High-Throughput Track Following Algorithm for Parallel Architectures
• A Downstream and vertexing algorithm for Long Lived Particles (LLP) selection at the first High-level trigger (HLT1) of LHCb
• An updated hybrid deep learning algorithm for identifying and locating primary vertices
• The SMARTHEP European Training Network
• Environmental sustainability in basic research: a perspective from HECAP+
• Optimizing Trigger-Level Track Reconstruction for Sensitivity to Exotic Signatures
• Transformers for Charged Particle Track Reconstruction in High Energy Physics
• An Automated Bandwidth Division for the LHCb Upgrade Trigger
• TrackFormers: In Search of Transformer-Based Particle Tracking for the High-Luminosity LHC Era
• Minimising Event Size, Maximising Physics: Inclusive Particle Isolation for LHCb's Run 3
• Energy efficiency of a GPU-based computing system for High Energy Physics experiments
• End-to-end multi-particle reconstruction in high occupancy imaging calorimeters with graph neural networks
• A parallel algorithm for fast reconstruction of primary vertices on heterogeneous architectures